Train Duel Deep Q Net using Pytorch on Gym Open AI environment. The model is generallized so that it can be used for any discrete-action-space environment on Gym Open AI by modifying in_features and out_features (number of actions) accordingly. Models can be saved and loaded for later traning sections.
References:
[1] Ralf C. Staudemeyer, and Eric Rothstein Morris. – Understanding LSTM –
a tutorial into Long Short-Term Memory Recurrent Neural Networks. In arXiv, September 2019.
[2] Ziyu Wang, Tom Schaul, Matteo Hessel, Hado van Hasselt, Marc Lanctot, and Nando de Freitas.
Dueling Network Architectures for Deep Reinforcement Learning. In arXiv, November 2015.
[3] Matthew Hausknecht, and Peter Stone. Deep Recurrent Q-Learning for Partially Observable
MDPs. In arXiv, July 2015.